4.1 Magnetism, Permeability, and Flux Leakage

The ASNT NDT Level II magnetic particle general exam, as administered prior to 15 December 2026, opens with Principles. Those items are not trivia about lodestones. They ask whether you can explain why particles collect at a crack, why they will not collect on aluminum, and what 'near saturation' actually means on a B-H curve.

Why flux leakage is the whole method

Magnetic particle testing (MT) finds discontinuities by making a leakage field visible. You magnetize a ferromagnetic part so that magnetic flux prefers to stay inside the metal. A surface or near-surface interruption forces some of that flux into the air. Fine ferromagnetic particles — dry powder or particles suspended in a liquid vehicle, colored or fluorescent — are attracted to that leakage field and pile up as an indication.

If there is no usable flux in the part, there is no leakage field. If the discontinuity does not interrupt the flux path, there is no useful leakage. Those two sentences decide almost every Principles item.

Industry practice is written around this physics. ASTM E709 (standard guide) and ASTM E1444 / E1444M (standard practice, widely used in aerospace), and ASME Boiler and Pressure Vessel Code, Section V, Article 7 for boilers and pressure vessels, all assume a ferromagnetic article and a field that can leak at the discontinuity. They do not invent a special rule that makes austenitic stainless steel suddenly magnetic. ASNT does not publish a secret permeability number for the written exam; the physics and those industry practices are what the general paper tests.

Ferromagnetic, paramagnetic, and diamagnetic

Exam stems sort materials into three magnetic classes. Only one class is an MT material.

Austenitic stainless steels such as Type 304 and Type 316 have a face-centered-cubic austenite structure. That structure does not support the domain alignment that makes carbon steel ferromagnetic. A shop magnet that 'sticks a little' to a heavily cold-worked 304 fitting, or to a duplex weld with some delta ferrite, does not convert the part into an MT article. The written exam still treats austenitic stainless, aluminum, and copper alloys as materials on which MT fails as a primary method.

Some 400-series stainless steels (ferritic and martensitic grades such as 410 and 420) are ferromagnetic. Read the alloy, not the word 'stainless.' A stem that says '410 stainless shaft' is an MT part; a stem that says '316L stainless nozzle' is not.

Do not confuse electrical conductivity with ferromagnetism. Aluminum and copper are excellent electrical conductors — they are eddy-current materials, not magnetic-particle materials. Eddy current testing (ET) is not one of the five methods on this ASNT NDT Level II written program. Mention it only as a contrast. It is not the answer that 'saves' MT on aluminum.

Domain theory at exam level

A ferromagnetic solid is divided into magnetic domains — microscopic regions in which atomic magnetic moments already point the same way. In the unmagnetized state the domains point in different directions and cancel, so the part has essentially no external field.

Apply an external magnetizing force H (ampere-turns per meter in SI; oersteds in older cgs exam language):

1. Domain walls move so favorably oriented domains grow.
2. Remaining domains rotate toward H.
3. At saturation, essentially all domains are aligned. Further increases in H add almost no flux density B (tesla in SI; gauss in cgs).

Remove H and some alignment remains if the material has retentivity. That leftover B is the residual field (remanence). The reverse H needed to drive B back to zero is the coercive force.

You do not need Bloch-wall mathematics. You do need this story: MT works because ferromagnetic domains can be lined up into a strong internal flux, and a crack is a low-permeability gap that those lined-up domains cannot jump without leaking. Paramagnetic and diamagnetic materials do not form these domains. There is nothing to line up.

Permeability, retentivity, residual field, and active field

Permeability μ is how readily the material accepts flux: qualitatively, B compared with H. High-μ steel is an easy magnetic path. Air, a crack, paint, and copper all have μ near that of free space. Flux would rather stay in steel than jump into a crack. When it is forced to jump, it leaks at the surface and particles can see it.

Retentivity is the material's ability to hold a residual field after the magnetizing current or yoke is turned off. Hardened steels and many tool steels are high-retentivity. Soft low-carbon steels, annealed weld metal, and many cast irons are lower-retentivity. Hard magnetic materials (high remanence, high coercivity) hold residual field and resist demagnetization. Soft magnetic materials magnetize and demagnetize easily and are poor residual-method parts.

Two operating modes follow from that distinction:

• An active field (also called the applied or continuous field) exists while the current is flowing or the yoke is energized. Particles are applied during magnetization. Sensitivity is highest because B is on the rising part of the magnetization curve, not merely at remanence.
• A residual field remains after the magnetizing force is removed. The residual method applies particles only after current-off. It works only when remanence is high enough to leak at the discontinuities of interest.

A Level II who residual-tests a soft, low-carbon weldment with alternating current (AC) is mixing two reasons the residual field is weak: the material will not hold much remanence, and AC reverses each half-cycle and leaves almost no residual. Residual techniques in production are typically direct current (DC) or rectified current on high-retentivity parts.

The B-H curve and hysteresis, qualitatively

Plot H on the horizontal axis and B on the vertical axis. Starting from a demagnetized part:

• B rises slowly at first (domain-wall pinning), then steeply (easy wall motion), then flattens at saturation.
• Reduce H to zero: B does not return to zero. It sits at remanence, Br. That point is the residual-method operating point.
• Reverse H until B = 0: that reverse field is coercivity, Hc.
• Push to reverse saturation and back, and you trace a hysteresis loop. The loop area is related to hysteresis loss; a wide loop is a hard, high-retentivity material.

Exam uses of the loop:

• Saturation versus leakage. Below saturation, flux can still detour around a tight crack inside the metal, so surface leakage is weak. Near saturation the material cannot accept much more flux, so a surface crack dumps flux into the air. That is why procedures talk about magnetizing 'to a level approaching saturation' rather than 'as much current as the station can put out.'
• Over-magnetization does not create more useful crack indications. Extra field paints the whole surface with background particles, buries contrast, and can make geometric corners look like cracks.
• Demagnetization is walking the loop down with a reversing, decreasing H until residual B is below the procedure limit — covered in the magnetization chapter. For Principles, know that a high-Hc material is harder to demagnetize and a better residual-method candidate.

Flux leakage at a surface or near-surface disruption

Picture flux lines running just under a ground weld toe. A surface-breaking fatigue crack is a thin air-filled slot. Flux lines that hit that slot cannot stay in steel; they bow out of the surface, loop through the air, and re-enter on the other side. Particles follow that external loop and form a sharp linear indication.

A near-surface slag stringer or subsurface crack does the same thing more weakly. Some flux still goes around the void inside the metal. The external leakage that reaches the surface is broader and fainter. Depth capability then depends on current type: AC stays at the skin; DC and half-wave direct current (HWDC) reach deeper. That depth ranking is developed with flux fields and discontinuity effects; the Principles point is simply that leakage at the surface is what the particles can see.

A crack that is parallel to the flux is almost invisible: the lines never have to cross the gap. Principles items and Flux Fields items share that orientation rule.

Why MT fails on austenitic stainless, aluminum, and copper

These materials cannot be driven to a ferromagnetic flux density. Their permeability is so close to air that a crack is not a meaningful change in magnetic path. You can clamp a yoke on a 316L nozzle all day; you will not get a valid leakage-field examination. The correct methods are visual testing (VT) plus liquid penetrant testing (PT), or a volumetric method if the concern is subsurface.

The same 'no' applies to copper bus bars, aluminum aerospace skins, titanium fittings, and nickel-base alloys that are not ferromagnetic. A faint magnetic pull after severe cold work is a materials curiosity, not a procedure qualification.

Realistic exam scenarios

A procedure writes fluorescent wet particles, continuous method, on a quenched-and-tempered 4340 landing-gear fitting. That material is ferromagnetic and high-retentivity. Continuous magnetization still gives the more sensitive active field; residual may be allowed later for process flow if the procedure proves residual field is adequate.

A Level II is handed a Type 304 stainless pipe spool and a yoke. The correct action is to stop and use a method the material supports. A faint pull on a pocket magnet after weld dressing is not a green light.

A carbon-steel shaft is magnetized well below the knee of the B-H curve. A known surface notch on a pie-field indicator or a quantitative quality indicator (QQI) barely shows. Increase current toward the procedure's magnetization level (near saturation for that technique) before you blame the particles.

A copper chill plate sitting next to a steel weldment picks up a few particles from overspray. That is not an indication. Copper cannot leak a ferromagnetic flux.

What Principles items are really testing

If the stem names the material, ask: ferromagnetic or not? If it names residual versus continuous, ask: is there enough remanence? If it names a missed crack on a steel part, ask: was there flux, and did the flux cross the crack? Those three questions are the Principles domain.